AIDS Behav DOI 10.1007/s10461-014-0948-z

ORIGINAL PAPER

Heroin Use and HIV Disease Progression: Results from a Pilot Study of a Russian Cohort E. Jennifer Edelman • Debbie M. Cheng • Evgeny M. Krupitsky • Carly Bridden • Emily Quinn • Alexander Y. Walley • Dmitry A. Lioznov Elena Blokhina • Edwin Zvartau • Jeffrey H. Samet

•

Ó Springer Science+Business Media New York 2014

Abstract Opioids have immunosuppressive properties, yet their impact on HIV disease progression remains unclear. Using longitudinal data from HIV-infected antiretroviral therapy-naı¨ve Russian individuals (n = 77), we conducted a pilot study to estimate the effect of heroin use on HIV disease progression. Heroin use was categorized based on past 30 days self-reported use at baseline, 6 and 12 months as none, intermittent or persistent. We estimated the effect of heroin use on HIV disease progression, measured as change in CD4 count from baseline to 12 months, using multivariable linear regression. Those with intermittent (n = 21) and no heroin use (n = 39) experienced

An earlier version of this work was presented at the 74th Annual Meeting of the College on Problems of Drug Dependence, Palm Springs, California, June, 2012. E. J. Edelman (&) Section of General Internal Medicine, Department of Medicine, Yale University School of Medicine, P.O. Box 208025, New Haven, CT 06520-8088, USA e-mail: [email protected]

mean decreases in CD4 count from baseline to 12 months (-103 and -10 cells/mm3, respectively; adjusted mean difference (AMD) -93; 95 % CI -245, 58). Those with persistent use (n = 17) showed a mean increase of 53 cells/mm3 (AMD 63; 95 % CI -95, 220). Future studies exploring the effects of heroin withdrawal on HIV disease progression are warranted.

Resumen Los opioides tienen propiedades inmunosupresoras, pero su impacto sobre la progresio´n de la enfermedad VIH sigue siendo poco clara. Utilizando datos longitudinales de infectados por el VIH terapia antirretroviral personas rusas (n = 77), se realizo´ un estudio piloto para estimar el efecto del uso de la heroı´na sobre la progresio´n de la enfermedad VIH. Uso de la heroı´na se clasifico´ E. Quinn Department of Biostatistics, Data Coordinating Center, Boston University School of Public Health, Boston, MA, USA

D. M. Cheng Department of Biostatistics, Boston University School of Public Health, Boston, MA, USA

A. Y. Walley
J. H. Samet Clinical Addiction Research and Education (CARE) Unit, Section of General Internal Medicine, Department of Medicine, Boston University School of Medicine/Boston Medical Center, Boston, MA, USA

E. M. Krupitsky
D. A. Lioznov
E. Blokhina
E. Zvartau First St. Petersburg Pavlov State Medical University, St. Petersburg, Russian Federation

J. H. Samet Department of Community Health Sciences, Boston University School of Public Health, Boston, MA, USA

E. M. Krupitsky St. Petersburg Bekhterev Research Psychoneurological Institute, St. Petersburg, Russian Federation C. Bridden Section of General Internal Medicine, Boston Medical Center, Department of Medicine, Clinical Addiction Research and Education (CARE) Unit, Boston, MA, USA

123

AIDS Behav

en u´ltimos 30 dı´as auto-informe en el momento de referencia, 6 meses y 12 meses como ninguno, intermitente o persistente. Se estimo´ el efecto del uso de la heroı´na sobre la progresio´n de la enfermedad VIH, medido como cambio de recuento de CD4 en la lı´nea de base para 12 meses, mediante regresio´n lineal multivariante. Aquellos con intermitente (n = 21) y no uso de la heroı´na (n = 39) experimentaron disminuciones promedio del nu´mero CD4 desde el nivel basal de 12 meses (-103 ce´lulas/mm3 y 10 ce´lulas/mm3, respectivamente; diferencia de medias ajustadas (AMD) -93; IC 95 % -245, 58). Las personas con uso persistente (n = 17) mostraron un aumento medio de 53 ce´lulas/mm3 (AMD 63; IC 95 % -95, 220). Futuros estudios que exploren los efectos de la heroı´na retirada sobre la progresio´n de la enfermedad VIH esta´n garantizados.

Keywords

Heroin
HIV disease progression
HIV
CD4

Introduction Globally, approximately 3 million people who inject drugs are estimated to be HIV-infected, with the highest numbers of injecting individuals in China, the United States and Russia [1]. Substance use impacts HIV disease progression by leading to decreased access and linkage to treatment [2], non-adherence to antiretroviral therapy (ART) [3, 4], and comorbidities [5]. Given heroin’s immunosuppressive properties, one could hypothesize that it may hasten HIV disease progression [6, 7]. The specific impact of heroin, a pro-drug of morphine, on HIV disease progression remains unclear [8, 9]. Laboratory studies demonstrate that morphine impacts multiple aspects of the immune system [10], including promoting HIV-1 growth in cell cultures [11], inducing lymphocyte apoptosis [12] and increasing expression of the HIV entry co-receptors with increased HIV replication [13–15]. Animal [16–19] and human [20–26] studies are inconsistent, with only some studies demonstrating deleterious effects of opioids on HIV progression [22–28]. Human studies evaluating the impact of heroin use on HIV progression have been limited. Few include longitudinal designs where frequency of heroin use is measured [8, 10, 20, 29], and heroin use is often categorized with cocaine use, a drug that has already been associated with HIV disease progression and has a completely different mechanism of action [8, 30, 31]. Moreover, studies are unable to adjust for important confounders, such as ART use and adherence [8, 32]. Therefore, we conducted the current pilot longitudinal study to examine the association between self-reported heroin use and HIV disease progression in a sample of HIV-infected Russians who were ART-naı¨ve. This represented a unique opportunity to examine these questions given that in Russia, a

123

cohort of HIV-infected persons not on ART is accessible, heroin is the dominant illicit substance used [33], cocaine use is quite limited [34, 35], and chronic opioids for pain or as opioid agonist therapy are not prescribed [36], thus minimizing issues of confounding. The primary objective of this study was to estimate the effect of heroin use on HIV disease progression as measured by change in CD4 cell count over a 12 month period.

Methods Study Design and Participants We conducted a longitudinal study to assess the effect of heroin use on change in CD4 cell counts in a prospective cohort of HIV-infected Russians. Participants were recruited between January 2010 and March 2011 for the Drug Use–– Investigating and Measuring Progression to AIDS in a Cohort Trial (IMPACT). This prospective cohort study in St. Petersburg was nested within the HIV Evolution in Russia–– Mitigating Infection Transmission and Alcoholism in a Growing Epidemic (HERMITAGE) trial [37]. The HERMITAGE study recruited 700 HIV-infected heavy drinkers who reported recent unprotected sex from four clinical inpatient and outpatient HIV and addiction sites in St. Petersburg, Russia, including (1) Botkin Infectious Disease Hospital, (2) the St. Petersburg AIDS Center, (3) Pavlov State Medical University Clinics, and (4) the St. Petersburg State Drug Treatment Clinic. Participants were also recruited from nonclinical sources including a needle exchange program, and through ‘‘snowball recruitment’’ and referred to one of the clinical sites for eligibility assessment. Eligibility criteria for the HERMITAGE study included the following: (1) HIVinfection; (2) aged 18–70 years old; (3) any at-risk drinking, as defined by the US National Institute on Alcohol Abuse and Alcoholism (NIAAA) [38], 6 months prior to hospital admission; (4) anal or vaginal sex without a condom during the past 6 months; (5) free from cognitive impairment or acute illness; (6) not trying to get [partner] pregnant; (7) provision of contact details for self and two confidants to facilitate followup; (8) living within 150 km of St. Petersburg; (9) no pending incarceration; and (10) able to provide informed consent. The IMPACT study recruited HERMITAGE subjects when they returned for their 12-month follow-up visit. To eliminate the effect of ART on the study outcome, change in CD4 cell count, the analytic sample for the current study was restricted to patients who were ART naı¨ve. Participants were also required to have CD4 cell count data available at both baseline and 12 months. The study was approved by the institutional review boards of Boston University and First St. Petersburg Pavlov State Medical University. All subjects provided written informed consent and were reimbursed the equivalent of

AIDS Behav

USD$10 for their completion of the additional specific IMPACT survey at the HERMITAGE follow-up visit. Assessments Data were collected for the HERMITAGE study at baseline, 6 and 12-month follow-up. For this analysis, data from the HERMITAGE study were supplemented with specific IMPACT data that were collected at the 12-month follow-up, including a survey, CD4 cell count and urine toxicology testing. Outcome Change in CD4 cell count (cells/mm3) from baseline to 12 months was the primary outcome. Baseline CD4 cell count was ascertained based on chart review, using the value available closest in timing to the baseline assessment. CD4 cell count at 12 months was directly measured using a standard assay (Fluorescence-activated cell sorting/Flow Cytometry). Primary Independent Variable: Pattern of Heroin Use Self-reported substance use was assessed using a modified risk behaviors survey (RBS) [39–41] at baseline, 6 and 12 months. Participants were asked specifically about use of heroin not in combination with other drugs and separately about their use of other opiates in the last 30 days. Based on responses across the three time points, the primary independent variable––pattern of heroin or other opioid use––was categorized into a 3-level variable as none (no reported heroin or other opioid use use at baseline, 6 and 12 months), intermittent (reported both use and no use across baseline, 6 and 12 months) or persistent (reported heroin or other opioid use at baseline, 6 and 12 months). Given that heroin is the opioid used by the majority of participants, we refer to the main independent variable as ‘‘pattern of heroin use’’ throughout the manuscript. Participants were categorized accordingly regardless of whether opioid use occurred alone or in combination with other drugs. There were no missing data on heroin use at baseline or 12 months. However, there were eight participants with missing data at 6 months. Of these eight participants, four reported use at baseline but no use at 12 months. These participants were classified as having intermittent use. For the remaining four participants (one who reported use at both baseline and 12 months and the other three who reported no use at both baseline and 12 months), we assumed use at 6 months was consistent with their reported use at baseline and 12 months. Given the variability in types of opioids used in Russia [42], we also assessed whether participants ever used heroin by itself, other opioids (codeine, china white, methadone, fentanyl) by themselves, or heroin mixed with

stimulants (Jeff, ephedrine, amphetamine, methamphetamine, cocaine or crack). In addition, we assessed the average number of injections and use without injection of heroin or other opioids in a given day over the prior 30 days at baseline, 6 and 12 months. To validate the selfreport data, we conducted urine toxicology tests at 12 months to test for opiates, which were conducted on the same day as the survey. Covariates Socio-demographic variables included gender and age. Depressive symptoms in the past 2 weeks was evaluated using the Beck Depression Inventory-II defined as a score greater than 13 [43] at baseline. Pain was assessed using the SF-12 [44] and dichotomized based on the presence of at least moderate pain at baseline, 6 or 12-months. Past 30 days stimulant (typically ephedrine in Russia) and cannabis use, assessed with a modified RBS [39, 40], and binge drinking [38], determined with the interview-based 30-days time-line follow back (TLFB) [45], were analyzed as dichotomous variables based on the presence of any use at baseline, 6 or 12-months. Binge drinking was defined as the consumption of 5 or more drinks on any given occasion for men and 4 or more for women. Time between baseline and follow-up CD4 cell counts was included given an expected time-dependent decline in CD4 cell count in one with untreated HIV and variability in the time between CD4 cell count assessments. Statistical Analysis We performed descriptive statistics to characterize the demographic and clinical characteristics of the analytic sample, overall and stratified by heroin use (none; intermittent; persistent). Baseline characteristics were compared across heroin use groups using one way-ANOVA for continuous variables and Chi square or Fisher’s exact test for categorical variables. We characterized opioid use among patients with intermittent or persistent use across all time points. Using urine toxicology results as the reference, we determined the sensitivity and specificity of selfreported opioid use. Correlation between variables was assessed using Spearman correlation (no pair of variables had r [ 0.50). Linear regression models were constructed to estimate the association between pattern of heroin use and change in CD4 cell count from baseline to 12 months. We conducted a series of adjusted analyses to control for potential confounding factors. The first model included the covariates gender, age, baseline CD4 cell count, time between baseline and CD4 cell counts, and depressive symptoms. Additional models were fit controlling for pain, stimulant use, cannabis use, and binge drinking in addition

123

AIDS Behav Table 1 IMPACT participant demographic and clinical characteristics (n = 77)

Characteristic

Pattern of heroin use over 12 months Overall

Age, years, mean (SD)

Persistent n = 17 (22 %)

p value*

30 (5)

29 (5)

30 (3)

31 (5)

0.48

25 (64 %)

10 (48 %)

11 (65 %)

0.41

45 (13, 91)

46 (10, 87)

65 (22, 95)

42 (23, 93)

0.77

Hepatitis C, n (%)

71 (92 %)

34 (87 %)

20 (95 %)

17 (100 %)

0.21

Depressive symptoms, n (%)

49 (64 %)

18 (46 %)

19 (91 %)

12 (71 %) \0.01

Baseline CD4 cells/mm3, mean (SD)

543 (274)

574 (211)

517 (314)

503 (351)

0.60

8 (10 %)

1 (3 %)

2 (10 %)

5 (29 %)

0.01

Months since 1st positive HIV test, median (IQR)

Stimulant use, n (%) Cannabis use, n (%)

14 (18 %)

5 (13 %)

4 (19 %)

5 (29 %)

0.33

Binge drinking, n (%)

67 (87 %)

32 (82 %)

19 (91 %)

16 (94 %)

0.40

Pain, n (%)

52 (68 %)

23 (59 %)

15 (71 %)

14 (82 %)

0.21

to the above covariates. Adjusted mean differences for persistent and intermittent groups versus no use are reported along with 95 % confidence intervals. Secondary, we ran a sensitivity analysis, excluding those four participants with missing data who either had reported use at both baseline and 12 months (n = 1) or had no reported use at both baseline and 12 months (n = 3). Third, analyses were performed categorizing heroin use as any versus none. Finally, confirmatory analyses were performed using median regression [46, 47] as this method does not rely on the normality assumption and is more robust to outliers than linear regression. We considered statistical significance to be a p \ 0.05. All analyses were performed using SAS version 9.3 (SAS Institute, Inc, NC, USA).

Results Participant Characteristics Among the 700 participants in the HERMITAGE study, 135 completed the IMPACT survey at 12 months. Of these, 53 were on ART and 5 had missing CD4 cell count data, leaving a final analytic sample of 77 individuals for the current study as described in Table 1. Heroin use was common with 27 % reporting intermittent use and 22 % reporting persistent use. Participant characteristics included the following: 40 % women; mean age of 30 years; diagnosed with HIV for less than 5 years on average; and a mean CD4 cell count of 543 cells/mm3. The mean baseline CD4 cell count was 574 cells/mm3 (standard deviation (SD) 211) among those with no heroin use, 517 cells/mm3 (SD 314) among those with intermittent use and 503 cells/ mm3 (SD 351) among those with persistent use (p = 0.60). The median duration between baseline and follow-up CD4

123

Intermittent n = 21 (27 %)

46 (60 %)

Male gender, n (%)

* p \ 0.05 considered to be statistically significant

None n = 39 (51 %)

cell counts was 14 months (interquartile range = 11,18). Depressive symptoms were the norm, particularly among those with intermittent (91 %) and persistent (71 %) heroin use. The majority reported pain (68 %). Stimulant use was also more prevalent among those with persistent heroin use (29 %) compared to those with intermittent (10 %) or no (3 %) heroin use. While cannabis use and binge drinking were more commonly reported among those with persistent heroin use than those with intermittent or no use, these differences were not statistically significant.

Self-Reported Heroin Use Versus Urine Toxicology Testing Across all time points, the majority of patients with either intermittent or persistent heroin use reported heroin use by itself (79 %); 18 % reported using other opioids and 3 % reported heroin used mixed with stimulants. Patients with intermittent or persistent heroin use predominantly reported use of heroin or other opioids by injection, with a median average number of injections per day of 2. Participants reported infrequent use of heroin or other opioids without injection (median average number of use without injection per day of 0; only one person (2.6 %) reported using heroin or other opioids without injection and had an average daily use greater than zero. Among the 24 participants who reported any opioid use in the past 30 days at 12 months, 23 were positive by urine toxicology testing. Among the 53 participants who reported no opioid use at 12 months, 3 were positive by urine toxicology testing; these 3 participants had all been classified as intermittent users based on self-report. The sensitivity of self-reported opioid use was 88.5 %; the specificity was 98.0 % using the urine tests as the gold standard.

AIDS Behav Table 2 Results from multiple linear regression models evaluating association between pattern of heroin use and changes in CD4 cell count (cells/mm3) from baseline to 12 months No heroin use

Intermittent heroin use

Persistent heroin use

Adjusted Means

Unadjusted model

-24

-93

52

Intermittent use versus no heroin use

Persistent use versus no heroin use

Adjusted mean difference (95 % CI)

p value

Adjusted mean difference (95 % CI)

p value

-69

0.38

76

0.37

(-225, 87) Adjusted modela

-10

-103

53

-93

Adjusted modela ? pain

-13

-106

47

-93

40

(-245, 58) -87

(-91, 244) 0.23

63

0.23

60

0.46

0.26

(-99, 218) 87

0.32

(-245, 58)

Adjusted modela ? stimulants

-47

-134

(-95, 220)

(-239, 65) Adjusted modela ? cannabis

-22

-114

46

-92

(-84, 258) 0.24

(-243, 60) Adjusted modela ? binge

-41

-135

17

-94

68

0.40

(-92, 229) 0.22

(-244, 57) a

0.44

58

0.47

(-99, 215)

Adjusted for age, gender, depressive symptoms, baseline CD4 cell count, time between baseline and 12 month CD4 cell counts (in days)

Table 3 (Sensitivity Analysis). Adjusted means and adjusted mean difference in CD4 cell count (cells/mm3) from baseline to 12 months by pattern of heroin use––excluding 4 subjects with missing 6-month heroin information No heroin use

Intermittent heroin use

Persistent heroin use

Adjusted means Unadjusted model Adjusted modela Adjusted Modela ? pain Adjusted modela ? stimulants Adjusted modela ? cannabis Adjusted modela ? binge

Intermittent use versus no heroin useb

Persistent use versus no heroin useb

Adjusted mean difference

Adjusted mean difference

95 % CI

95 % CI

-12

-93

56

-81

-242, 81

69

-108, 246

4

-125

50

-129

-288, 29

46

-119, 211

-33

-156

38

-129

-287, 29

44

-121, 210

8

-122

52

-123

-281, 36

71

-108, 250

-29

-160

11

-130

-290, 29

44

-125, 212

2

-127

46

-131

-288, 26

40

-124, 204

a

Adjusted for age, gender, depressive symptoms, baseline CD4 cell count, time between baseline and 12 month CD4 cell counts (in days)

b

No values reached statistical significance defined as a p \ 0.05

Heroin Use and HIV Disease Progression In the unadjusted analyses, those with intermittent and no heroin use experienced mean decreases in CD4 cell count from baseline to 12 months (-93 and -24 CD4 cells/ mm3, respectively; mean difference -69; 95 %CI -225, 87) (Table 2). Those with persistent heroin use experienced a mean increase of 52 cells/mm3 (mean difference 76; 95 %CI -91, 244). The results were similar after adjusting for gender, age, baseline CD4 cell count, time between CD4 cell counts and

depressive symptoms. Those with intermittent and no heroin use (comparison group) experienced mean decreases in CD4 cell count from baseline to 12 months (-103 and -10 cells/mm3, respectively; adjusted mean difference (AMD) -93; 95 % CI -245, 58). Those with persistent heroin use showed a mean increase of 53 cells/mm3 compared to those with no use (AMD 63; 95 %CI -95, 220). After including pain in the model, the results were similar (AMD -93; 95 %CI -245, 58 for intermittent versus no heroin use; AMD 60; 95 %CI -99, 218 for persistent versus no heroin use). These patterns persisted in the

123

AIDS Behav

models including stimulants, cannabis and binge drinking. In sensitivity analysis, excluding the one participant with reported use at both baseline and 12 months and the three who reported no use at both baseline and 12 months, findings were consistent (Table 3). Analyses categorizing heroin use as any versus none and confirmatory analyses using median regression models resulted in similar findings in terms of the directions of observed associations and lack of statistical significance (data not shown).

Discussion Among this sample of young, HIV-infected Russians with risky drinking, the magnitude and direction of the estimated association between heroin use and CD4 cell count change appeared to depend on whether the heroin use was intermittent or persistent. Although as anticipated in this pilot study with none of the observed differences being statistically significant, those with intermittent heroin use appeared to have larger decreases in CD4 cell count between baseline and 12 months compared to those with no heroin use, while those with persistent heroin use appeared to have an increase in CD4 cell count over time. There are very limited human data examining the longitudinal association between CD4 cell counts and different patterns of heroin use among a sample naı¨ve to ART and not currently using other non-heroin opioids. The current study seeks to examine this particular issue by including repeated measures of use of individual substances in a distinctive cohort who are naı¨ve to ART and not using other opioids and relatively recently HIV-infected as suggested by the mean CD4 cell count. These analyses also include relevant covariates known to impact HIV disease progression (e.g. baseline CD4 cell count, depressive symptoms, stimulants). Given the absence of opioid agonist therapy in Russia, intermittent opioid use is associated with recurrent withdrawal episodes. Thus, the observed decreases in CD4 cell count for intermittent heroin use compared to no heroin use and the observed increases in CD4 cell count for persistent heroin use support the notion that opioid withdrawal may be particularly harmful to the immune system while chronic opioid use may be associated with some degree of immune tolerance [7, 16, 48]. Preclinical studies support such a hypothesis. For example, one study of opioid dependent macaques found that stable opioid dependence was not associated with a decline in CD4 cells in comparison to the pre-morphine condition [49]. Precipitated opioid withdrawal or abstinence for 24 h, however, was associated with a decrease in absolute lymphocyte counts as well as lymphocyte subsets, including CD4 cells, in comparison to both the pre-morphine and opioid-dependent

123

states [49]. In addition, pilot data indicate that precipitated opioid withdrawal in monkeys led to a rise in circulating cells with simian immunodeficiency virus indicating that withdrawal may enhance viral replication [16]. Data from humans are mixed. One study of HIV-uninfected heroin users revealed that active use and early withdrawal (i.e., 15–21 days) was associated with a lower percentage and absolute count of CD4 cells in comparison to normal controls and those in late withdrawal (i.e., 6–24 months). Notably, those in early withdrawal had the lowest values though these differences were not statistically significant [50]. Data from the AIDS link to intravenous experiences (ALIVE) cohort, which had careful ascertainment of injection drug use, did not find a significant impact of injection drug use pattern on CD4 cell count decline [51]. One reason our findings may differ from this study is their inclusion of a sicker population indicated by the fact that many of these patients developed AIDS and the fact that the issue examined was injection drug use rather than heroin use. Also, the ALIVE study was unable to focus on heroin use in the absence of other opioid use such as pain meds or opioid agonist treatment. In contrast, a study of the women’s interagency HIV study (WIHS) cohort demonstrated that progression to AIDS was greater among those with consistent, inconsistent and former drug use compared to never users [29]. Lucas et al. found that compared to nonusers, those with recent active use in the setting of intermittent use (AOR = 2.3, 95 % 1.5, 3.0), and those with persistent use (AOR = 2.1, 95 % 1.4, 3.1) were more likely to develop an opportunistic infection. Compared to nonusers, however, those with recent abstinence in the setting of intermittent use (AOR = 1.4, 95 % 1.0, 1.9) were not significantly more likely to develop an opportunistic infection [20]. Both of these studies included patients receiving ART and categorized heroin with other substances including stimulants limiting interpretation of the findings [20, 29]. We specifically excluded patients on ART to avoid masking any effects of opioids on HIV progression in the setting of ART. While the precise mechanisms mediating the effects of opioids on the immune system remain to be elucidated, they are likely to involve both the hypothalamic–pituitary-adrenal axis as well as direct effects on various immune cells [6, 7, 48, 49, 52]. Moreover, intermittent opioid use may reflect financial insecurity, which may be associated with food insecurity and poor nutrition, and subsequent immune dysfunction. We assessed whether pain symptoms impacted the association between heroin use and change in CD4 cell count because pain and heroin use are associated and pain as a source of stress may impact CD4 cell count; however, our findings persisted across these models [53]. Similarly, as depression has been found to be associated with HIV disease progression specifically among HIV-infected

AIDS Behav

patients with injection drug use [54] we attempted to adjust for depressive symptoms in our models. While our study was not designed to specifically evaluate the impact of pain or depressive symptoms on HIV disease progression, prior work indicates the prevalence and importance of addressing these symptoms in this population [53, 55]. Stimulant [30, 31] and alcohol use [56, 57] have been reported to have deleterious effects on HIV disease progression with less clear effects of cannabis on HIV disease progression [58]. While the presence of any stimulant use varied across patterns of heroin use, the overall amount of use based on days of use was low across groups (data not shown). Again, results did not change after adjusting for use of these other substances. Consistent with our inclusion criteria, binge drinking was common across all categories of heroin use. Alcohol may be used in combination with opioids to moderate a high or to cope with withdrawal symptoms. Whether the timing of binge drinking relative to opioid use leads to differential effects on HIV disease progression warrants further investigation. These findings should be interpreted in the context of some limitations. We lacked data on HIV viral load, which may have been a confounder. We relied on self-reported substance use, which may be subject to under-reporting. However, urine toxicology results were obtained to assess reliability of self-report and were found to be largely consistent with the self-report data. As the majority of the subjects in our sample were co-infected with HCV, we were unable to assess whether the effect of opioids on HIV disease progression varied by HCV status. Also, a single laboratory using consistent assays was not possible for the measurement of both the baseline and 12-month CD4 cell counts, with the possible consequence of decreasing the precision of these findings. We did not specifically ask about krokodil [42]. This opioid drug, however, has only been noted in St. Petersburg in recent years, after our study recruitment occurred. Finally, while the immunosuppressive effects of opioids may vary based on specific type of opioid [15, 52] and route of administration, our study was able to evaluate those effects mainly associated with injection use of heroin. These limitations notwithstanding, these data support the need for further investigation of the effects of opioids and different patterns of heroin use on HIV disease progression. In particular, larger future studies assessing the differential effects of chronic use, intermittent use and the associated withdrawal, compared to no heroin use, might be quite informative. In addition, whether the effects of heroin on HIV disease progression differ based on method of administration also deserves further evaluation. Despite the long-term recognition of the association between heroin use and HIV infection, important unanswered questions

about the nature of the relationship in terms of HIV disease progression remain. Acknowledgments The project was supported by the National Institute of Drug Abuse (R21 DA025435; R25-DA13582; and K12DA033312-01A1) and the National Institute on Alcohol Abuse and Alcoholism (K24 AA015674; R01 AA016059; U24AA020778; and U24AA020779). Conflict of interest

The authors have no known conflicts of interest.

References 1. Mathers BM, Degenhardt L, Phillips B, Wiessing L, Hickman M, Strathdee SA, et al. Global epidemiology of injecting drug use and HIV among people who inject drugs: a systematic review. Lancet. 2008;372(9651):1733–45. doi:10.1016/S0140-6736(08) 61311-2. 2. McGowan CC, Weinstein DD, Samenow CP, Stinnette SE, Barkanic G, Rebeiro PF, et al. Drug use and receipt of highly active antiretroviral therapy among HIV-infected persons in two U.S. clinic cohorts. PLoS One. 2011;6(4):e18462. doi:10.1371/ journal.pone.0018462. 3. Cofrancesco J Jr, Scherzer R, Tien PC, Gibert CL, Southwell H, Sidney S, et al. Illicit drug use and HIV treatment outcomes in a US cohort. AIDS. 2008;22(3):357–65. doi:10.1097/QAD.0b013e32 82f3cc21. 4. Kerr T, Marshall BD, Milloy MJ, Zhang R, Guillemi S, Montaner JS, et al. Patterns of heroin and cocaine injection and plasma HIV-1 RNA suppression among a long-term cohort of injection drug users. Drug Alcohol Depend. 2012;124(1–2):108–12. doi:10.1016/j.drugalcdep.2011.12.019. 5. Samet JH, Walley AY, Bridden C. Illicit drugs, alcohol, and addiction in human immunodeficiency virus. Panminerva Med. 2007;49(2):67–77. 6. McCarthy L, Wetzel M, Sliker JK, Eisenstein TK, Rogers TJ. Opioids, opioid receptors, and the immune response. Drug Alcohol Depend. 2001;62(2):111–23. 7. Roy S, Ninkovic J, Banerjee S, Charboneau RG, Das S, Dutta R, et al. Opioid drug abuse and modulation of immune function: consequences in the susceptibility to opportunistic infections. J Neuroimmune Pharmacol. 2011;6(4):442–65. doi:10.1007/ s11481-011-9292-5. 8. Kipp AM, Desruisseau AJ, Qian HZ. Non-injection drug use and HIV disease progression in the era of combination antiretroviral therapy. J Subst Abuse Treat. 2011;40(4):386–96. doi:10.1016/j. jsat.2011.01.001. 9. Cabral GA. Drugs of abuse, immune modulation, and AIDS. J Neuroimmune Pharmacol. 2006;1(3):280–95. doi:10.1007/ s11481-006-9023-5. 10. Kapadia F, Vlahov D, Donahoe RM, Friedland G. The role of substance abuse in HIV disease progression: reconciling differences from laboratory and epidemiologic investigations. Clin Infect Dis. 2005;41(7):1027–34. 11. Peterson PK, Sharp BM, Gekker G, Portoghese PS, Sannerud K, Balfour HH Jr. Morphine promotes the growth of HIV-1 in human peripheral blood mononuclear cell cocultures. AIDS. 1990;4(9):869–73. 12. Moorman J, Zhang Y, Liu B, LeSage G, Chen Y, Stuart C, et al. HIV-1 gp120 primes lymphocytes for opioid-induced, beta-arrestin 2-dependent apoptosis. Biochim Biophys Acta. 2009; 1793(8):1366–71.

123

AIDS Behav 13. Li Y, Merrill JD, Mooney K, Song L, Wang X, Guo CJ, et al. Morphine enhances HIV infection of neonatal macrophages. Pediatr Res. 2003;54(2):282–8. doi:10.1203/01.PDR.00000749 73.83826.4C. 14. Donahoe RM, Vlahov D. Opiates as potential cofactors in progression of HIV-1 infections to AIDS. J Neuroimmunol. 1998; 83(1–2):77–87. 15. Sacerdote P, Franchi S, Panerai AE. Non-analgesic effects of opioids: mechanisms and potential clinical relevance of opioid-induced immunodepression. Curr Pharm Des. 2012;18(37):6034–42. 16. Donahoe RM. Multiple ways that drug abuse might influence AIDS progression: clues from a monkey model. J Neuroimmunol. 2004; 147(1–2):28–32. 17. Kumar R, Torres C, Yamamura Y, Rodriguez I, Martinez M, Staprans S, et al. Modulation by morphine of viral set point in rhesus macaques infected with simian immunodeficiency virus and simian-human immunodeficiency virus. J Virol. 2004;78(20):11425–8. 18. Donahoe RM, O’Neil SP, Marsteller FA, Novembre FJ, Anderson DC, Lankford-Turner P, et al. Probable deceleration of progression of Simian AIDS affected by opiate dependency: studies with a rhesus macaque/SIVsmm9 model. J Acquir Immune Defic Syndr. 2009;50(3):241–9. 19. Chuang RY, Suzuki S, Chuang TK, Miyagi T, Chuang LF, Doi RH. Opioids and the progression of simian AIDS. Front Biosci. 2005;10:1666–77. 20. Lucas GM, Griswold M, Gebo KA, Keruly J, Chaisson RE, Moore RD. Illicit drug use and HIV-1 disease progression: a longitudinal study in the era of highly active antiretroviral therapy. Am J Epidemiol. 2006;163(5):412–20. 21. Lucas GM, Cheever LW, Chaisson RE, Moore RD. Detrimental effects of continued illicit drug use on the treatment of HIV-1 infection. J Acquir Immune Defic Syndr. 2001;27(3):251–9. 22. Poundstone KE, Chaisson RE, Moore RD. Differences in HIV disease progression by injection drug use and by sex in the era of highly active antiretroviral therapy. AIDS. 2001;15(9):1115–23. 23. Perez-Hoyos S, del Amo J, Muga R, del Romero J, de Garcia Olalla P, Guerrero R, et al. Effectiveness of highly active antiretroviral therapy in Spanish cohorts of HIV seroconverters: differences by transmission category. AIDS. 2003;17(3):353–9. 24. Moore RD, Keruly JC, Chaisson RE. Differences in HIV disease progression by injecting drug use in HIV-infected persons in care. J Acquir Immune Defic Syndr. 2004;35(1):46–51. 25. Egger M, May M, Chene G, Phillips AN, Ledergerber B, Dabis F, et al. Prognosis of HIV-1-infected patients starting highly active antiretroviral therapy: a collaborative analysis of prospective studies. Lancet. 2002;360(9327):119–29. 26. Thorpe LE, Frederick M, Pitt J, Cheng I, Watts DH, Buschur S, et al. Effect of hard-drug use on CD4 cell percentage, HIV RNA level, and progression to AIDS-defining class C events among HIV-infected women. J Acquir Immune Defic Syndr. 2004;37(3): 1423–30. 27. Mocroft A, Madge S, Johnson AM, Lazzarin A, Clumeck N, Goebel FD, et al. A comparison of exposure groups in the EuroSIDA study: starting highly active antiretroviral therapy (HAART), response to HAART, and survival. J Acquir Immune Defic Syndr. 1999;22(4):369–78. 28. Junghans C, Low N, Chan P, Witschi A, Vernazza P, Egger M. Uniform risk of clinical progression despite differences in utilization of highly active antiretroviral therapy: Swiss HIV Cohort Study. AIDS. 1999;13(18):2547–54. 29. Kapadia F, Cook JA, Cohen MH, Sohler N, Kovacs A, Greenblatt RM, et al. The relationship between non-injection drug use behaviors on progression to AIDS and death in a cohort of HIV seropositive women in the era of highly active antiretroviral

123

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43. 44.

therapy use. Addiction. 2005;100(7):990–1002. doi:10.1111/j. 1360-0443.2005.01098.x. Baum MK, Rafie C, Lai S, Sales S, Page B, Campa A. Crackcocaine use accelerates HIV disease progression in a cohort of HIV-positive drug users. J Acquir Immune Defic Syndr. 2009;50(1):93–9. Cook JA, Burke-Miller JK, Cohen MH, Cook RL, Vlahov D, Wilson TE, et al. Crack cocaine, disease progression, and mortality in a multicenter cohort of HIV-1 positive women. AIDS. 2008;22(11):1355–63. Palepu A, Tyndall M, Yip B, O’Shaughnessy MV, Hogg RS, Montaner JS. Impaired virologic response to highly active antiretroviral therapy associated with ongoing injection drug use. J Acquir Immune Defic Syndr. 2003;32(5):522–6. Krupitsky EM, Zvartau EE, Lioznov DA, Tsoy MV, Egorova VY, Belyaeva TV, et al. Co-morbidity of infectious and addictive diseases in St. Petersburg and the Leningrad Region, Russia. Eur Addict Res. 2006;12(1):12–9. doi:10.1159/000088578. Krupitsky E, Zvartau E, Karandashova G, Horton NJ, Schoolwerth KR, Bryant K, et al. The onset of HIV infection in the Leningrad region of Russia: a focus on drug and alcohol dependence. HIV Med. 2004;5(1):30–3. Long EF, Brandeau ML, Galvin CM, Vinichenko T, Tole SP, Schwartz A, et al. Effectiveness and cost-effectiveness of strategies to expand antiretroviral therapy in St. Petersburg, Russia. AIDS. 2006;20(17):2207–15. doi:10.1097/QAD.0b013e32801 0c7d0. Cherny NI, Baselga J, de Conno F, Radbruch L. Formulary availability and regulatory barriers to accessibility of opioids for cancer pain in Europe: a report from the ESMO/EAPC Opioid Policy Initiative. Ann Oncol. 2010;21(3):615–26. doi:10.1093/ annonc/mdp581. Samet JH, Raj A, Cheng DM, Blokhina E, Bridden C, Chaisson CE, et al. HERMITAGE––a randomized controlled trial to reduce sexually transmitted infections and HIV-risk behaviors among HIV-infected Russian drinkers. Addiction. 2014;. doi:10.1111/ add.12716. National Institute on Alcohol Abuse and Alcoholism. What’s ‘‘atrisk’’ or ‘‘heavy’’ drinking? http://rethinkingdrinking.niaaa.nih. gov/IsYourDrinkingPatternRisky/WhatsAtRiskOrHeavyDrink ing.asp. Accessed 7.18 2011. Weatherby N, Needle R, Cesari H, Booth R, McCoy CB, Watters JK, Williams M, Chitwood DD. Validity of Self-Reported Drug Use among Injection Drug Users and Crack Cocaine Users Recruited through Street Outreach. Eval Program Plan. 1994;17(4):347–55. Dowling-Guyer S, Johnson M, Fisher D, Needle R, Watters J, Anderson M, Williams M, Kotransld L, Booth R, Rhodes E, Weatherby N, Estada A, Fleming D, Deren S, Tortu S. Reliability of drug users’ self-reported HIV risk behaviors and validity of self-reported recent drug use. Assessment. 1994;1(4):1383–92. Tyurina A, Krupitsky E, Cheng DM, Coleman SM, Walley AY, Bridden C, et al. Is cannabis use associated with HIV drug and sex risk behaviors among Russian HIV-infected risky drinkers? Drug Alcohol Depend. 2013;132(1–2):74–80. doi:10.1016/j.dru galcdep.2013.01.009. Grund JP, Latypov A, Harris M. Breaking worse: the emergence of krokodil and excessive injuries among people who inject drugs in Eurasia. Int J Drug Policy. 2013;24(4):265–74. doi:10.1016/j. drugpo.2013.04.007. Beck AT. Depression Inventory. Russian Translation ed. Pearson Educatoin, Inc.; 1996, 2007. Ware J Jr, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220–33.

AIDS Behav 45. Sobell LC, Sobell SM. Alcohol Timeline Followback (TLFB); Handbook of Psychiatric Measures. Washington, D.C: American Psychiatric Association; 1996. 46. Hao L, Naiman DQ. Quantile regression. Thousand Oaks: Sage Publications; 2007. 47. Koenker R. Quantile regression. Cambridge; New York: Cambridge University Press; 2005. 48. Eisenstein TK, Rahim RT, Feng P, Thingalaya NK, Meissler JJ. Effects of opioid tolerance and withdrawal on the immune system. J Neuroimmune Pharmacol. 2006;1(3):237–49. doi:10.1007/ s11481-006-9019-1. 49. Weed MR, Carruth LM, Adams RJ, Ator NA, Hienz RD. Morphine withdrawal dramatically reduces lymphocytes in morphinedependent macaques. J Neuroimmune Pharmacol. 2006;1(3): 250–9. doi:10.1007/s11481-006-9029-z. 50. Govitrapong P, Suttitum T, Kotchabhakdi N, Uneklabh T. Alterations of immune functions in heroin addicts and heroin withdrawal subjects. J Pharmacol Exp Ther. 1998;286(2):883–9. 51. Lyles CM, Margolick JB, Astemborski J, Graham NM, Anthony JC, Hoover DR, et al. The influence of drug use patterns on the rate of CD4? lymphocyte decline among HIV-1-infected injecting drug users. AIDS. 1997;11(10):1255–62. 52. Sacerdote P. Opioid-induced immunosuppression. Curr Opin Support Palliat Care. 2008;2(1):14–8. doi:10.1097/SPC.0b013e32 82f5272e.

53. Tsui JI, Cheng DM, Coleman SM, Blokhina E, Bridden C, Krupitsky E, et al. Pain is associated with heroin use over time in HIV-infected Russian drinkers. Addiction. 2013;108(10):1779– 87. doi:10.1111/add.12274. 54. Bouhnik AD, Preau M, Vincent E, Carrieri MP, Gallais H, Lepeu G, et al. Depression and clinical progression in HIV-infected drug users treated with highly active antiretroviral therapy. Antivir Ther. 2005;10(1):53–61. 55. Palfai TP, Cheng DM, Coleman SM, Bridden C, Krupitsky E, Samet JH. The influence of depressive symptoms on alcohol use among HIV-infected Russian drinkers. Drug Alcohol Depend. 2014;134:85–91. doi:10.1016/j.drugalcdep.2013.09.014. 56. Baum MK, Rafie C, Lai S, Sales S, Page JB, Campa A. Alcohol use accelerates HIV disease progression. AIDS Res Hum Retroviruses. 2010;26(5):511–8. doi:10.1089/aid.2009.0211. 57. Samet JH, Cheng DM, Libman H, Nunes DP, Alperen JK, Saitz R. Alcohol consumption and HIV disease progression. J Acquir Immune Defic Syndr. 2007;46(2):194–9. doi:10.1097/QAI. 0b013e318142aabb. 58. Molina PE, Amedee A, LeCapitaine NJ, Zabaleta J, Mohan M, Winsauer P, et al. Cannabinoid neuroimmune modulation of SIV disease. J Neuroimmune Pharmacol. 2011;6(4):516–27. doi:10. 1007/s11481-011-9301-8.

123